Quality prediction device and method therefor, non-transitory computer readable medium storing program
A quality prediction device includes a probabilistic prediction model generation unit that generates a probabilistic prediction model on the basis of a plurality of training data in which a log relating to a molding operating condition or to a state of a molding machine and a quality value of a molding product corresponding to the log are associated with each other, and a quality prediction unit that calculates a quality index of the molding product from the log using the probabilistic prediction model.
1 . A quality prediction device comprising:
a processor configured to execute
generating of a probabilistic prediction model on the basis of a plurality of training data in which a log relating to a molding operating condition or to a state of a molding machine and a quality value of a molding product corresponding to the log are associated with each other,
setting a determination parameter, the determination parameter being designated by a user or adjusted according to a simulation result of the processor,
performing an injection molding and outputting the log from an injection molding machine body,
predicting a quality index of the molding product from the log using the probabilistic prediction model, and
making a determination whether inspection is required by comparing the predicted quality index to the determination parameter.
2 . The quality prediction device according to claim 1 ,
wherein the quality index indicates at least one of an expected value and a variance of the molding product, a probability of occurrence of a desired quality of the molding product, and a probability density distribution of a quality of the molding product.
3 . The quality prediction device according to claim 1 ,
wherein the determination parameter is adjustable.
4 . The quality prediction device according to claim 1 ,
wherein the probabilistic prediction model is updated according to newly acquired training data.
5 . The quality prediction device according to claim 1 , wherein the processor is further configured to execute:
obtaining and registering the quality value associated with the log in response to the determination that the inspection is required, and
performing next injection molding when a predetermined condition is met in response to the determination that the inspection is not required.
6 . The quality prediction device according to claim 1 , wherein the processor is further configured to execute:
notifying the determination whether the inspection is required to a user.
7 . A quality prediction method comprising:
generating a probabilistic prediction model on the basis of a plurality of training data in which a log relating to a molding operating condition or to a state of a molding machine and a quality value of a molding product corresponding to the log are associated with each other; and
setting a determination parameter, the determination parameter being designated by a user or adjusted according to a simulation result,
performing an injection molding and outputting the log from an injection molding machine body,
predicting a quality index of the molding product from the log using the probabilistic prediction model, and
making a determination whether inspection is required by comparing the predicted quality index to the determination parameter.
8 . The quality prediction method of claim 7 , further comprising:
obtaining and registering the quality value associated with the log in response to the determination that the inspection is required, and
performing next injection molding when a predetermined condition is met in response to the determination that the inspection is not required.
9 . The quality prediction method of claim 7 , further comprising:
notifying the determination whether the inspection is required to a user.
10 . A non-transitory computer readable medium storing a program that causes a computer to execute a process, the process comprising:
generating a probabilistic prediction model on the basis of a plurality of training data in which a log relating to a molding operating condition or to a state of a molding machine and a quality value of a molding product corresponding to the log are associated with each other; and
setting a determination parameter, the determination parameter being designated by a user or adjusted according to a simulation result,
performing an injection molding and outputting the log from an injection molding machine body,
predicting a quality index of the molding product from the log using the probabilistic prediction model, and
making a determination whether inspection is required by comparing the predicted quality index to the determination parameter.
11 . The non-transitory computer readable medium of claim 10 , wherein the process further comprises:
obtaining and registering the quality value associated with the log in response to the determination that the inspection is required, and
performing next injection molding when a predetermined condition is met in response to the determination that the inspection is not required.
12 . The non-transitory computer readable medium of claim 10 , wherein the process further comprises:
notifying the determination whether the inspection is required to a user.